# NVIDIA RTX PRO 6000 instead of H200 for ZeroGPU **URL:** \ **Category:** Spaces\ **Created:** [May 13, 2026, 12:14am UTC](https://discuss.huggingface.co/t/nvidia-rtx-pro-6000-instead-of-h200-for-zerogpu/175960 "2026-05-13T00:14:01Z")\ **Posts on this page:** 9\ **Page:** 1
**Author:** ![alexnasa](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/alexnasa/32/59701_2.png) [@alexnasa](https://discuss.huggingface.co/u/alexnasa)\ **Post date:** [May 13, 2026, 12:14am UTC](https://discuss.huggingface.co/t/nvidia-rtx-pro-6000-instead-of-h200-for-zerogpu/175960/1 "2026-05-13T00:14:01Z")
Spaces seems upon restarting today switch to NVIDIA RTX PRO 6000 for ZeroGPU spaces instead of running on H200 with lower VRAM, which breaks some of the Spaces from running properly > **[Ovi \[local\] - a Hugging Face Space by alexnasa](https://huggingface.co/spaces/alexnasa/Ovi-ZEROGPU)** > > Generate Hollywood Style Actors on your Local Machine GPU: NVIDIA RTX PRO 6000 Blackwell Server Edition Capability: (12, 0) CUDA: 12.8 ---
**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](https://discuss.huggingface.co/u/John6666)\ **Post date:** [May 13, 2026, 4:50am UTC](https://discuss.huggingface.co/t/nvidia-rtx-pro-6000-instead-of-h200-for-zerogpu/175960/2 "2026-05-13T04:50:53Z")
Yeah. I encountered the same issue in one of my Spaces. Since the NVIDIA RTX PRO 6000 is a Blackwell architecture GPU, I had to update PyTorch to version 2.7.1 or later and use cu128 or later, so I did just that. This might be difficult to resolve in Spaces where updates have stopped or that require an older version of PyTorch… Part of container log: ```auto /usr/local/lib/python3.10/site-packages/torch/cuda/ __init__.py:235: UserWarning: NVIDIA RTX PRO 6000 Blackwell Server Edition MIG 2g.48gb with CUDA capability sm_120 is not compatible with the current PyTorch installation. The current PyTorch install supports CUDA capabilities sm_50 sm_60 sm_70 sm_75 sm_80 sm_86 sm_90. If you want to use the NVIDIA RTX PRO 6000 Blackwell Server Edition MIG 2g.48gb GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/ ``` Edit: Oh… Seems `2.8.0` or later is necessary now… ```auto configuration error torch version in requirements.txt is not compatible with ZeroGPU. Supported versions: 2.11.0, 2.10.0, 2.9.1, 2.8.0 ``` ---
**Author:** ![alexnasa](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/alexnasa/32/59701_2.png) [@alexnasa](https://discuss.huggingface.co/u/alexnasa)\ **Post date:** [May 13, 2026, 9:40am UTC](https://discuss.huggingface.co/t/nvidia-rtx-pro-6000-instead-of-h200-for-zerogpu/175960/3 "2026-05-13T09:40:06Z")
It goes for all my other Spaces as well, it’s an issue across the board, which I managed to solve some of them but downgrading the Spaces to slower or lower resolution inferences, it just seems backwards to downgrade all Spaces to PRO 6000 from H200, is there no way to enforce H200 for ZeroGPU or is that a permanent downgrade for all Spaces that will happen now? Also upgrading to 2.10, Still gives me PRO 6000 as the GPU and not H200 2.10.0+cu128 12.8 NVIDIA RTX PRO 6000 Blackwell Server Edition (12, 0) 3.10.13 (main, Mar 12 2024, 12:16:25) [GCC 12.2.0] ---
**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](https://discuss.huggingface.co/u/John6666)\ **Post date:** [May 13, 2026, 10:17am UTC](https://discuss.huggingface.co/t/nvidia-rtx-pro-6000-instead-of-h200-for-zerogpu/175960/4 "2026-05-13T10:17:46Z")
It isn’t mentioned in the documentation, and I don’t know anything for sure… These are just lessons learned from observing how it works… > Does upgrading to Torch 2.8.0+ bring H200 support back? No. This is merely a workaround to prevent errors from occurring even if you happen to get a Blackwell-generation GPU. Edit: For example, many LTX-related Spaces use `torch==2.8.0`, but there is a possibility that they are causing generation errors due to insufficient VRAM: [LTX 2.3 Spaces Not Working!](https://discuss.huggingface.co/t/ltx-2-3-spaces-not-working/175961) ---
**Author:** ![alexnasa](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/alexnasa/32/59701_2.png) [@alexnasa](https://discuss.huggingface.co/u/alexnasa)\ **Post date:** [May 13, 2026, 10:48am UTC](https://discuss.huggingface.co/t/nvidia-rtx-pro-6000-instead-of-h200-for-zerogpu/175960/5 "2026-05-13T10:48:31Z")
RTX PRO 6000 has lower VRAM especially with the ZeroGPU overhead in comparison to H200, hence majority of big models that were running on H200 before wouldn’t be easily run, unless some offloading gets engineered in the code, in turn means more time for inference on the ZeroGPU quota ---
**Author:** ![hysts](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/hysts/32/32230_2.png) [@hysts](https://discuss.huggingface.co/u/hysts)\ **Post date:** [May 13, 2026, 10:59am UTC](https://discuss.huggingface.co/t/nvidia-rtx-pro-6000-instead-of-h200-for-zerogpu/175960/6 "2026-05-13T10:59:23Z")
Just a heads-up. The ZeroGPU docs are being updated to reflect this hardware change. > > > \> \[!NOTE\] > \> \*\*Low Risk\*\* > \> Documentation-only updates, but they change published… hardware specs and quota numbers, so incorrect values could mislead users if not aligned with product reality. > \> > \> \*\*Overview\*\* > \> Updates ZeroGPU documentation to reflect a \*\*hardware refresh from NVIDIA H200 to NVIDIA RTX Pro 6000 Blackwell\*\*, including updated VRAM figures and \`large\`/\`xlarge\` sizing details. > \> > \> Refreshes ZeroGPU quota/plan messaging across Hub docs (Free/PRO/Team/Enterprise daily minutes) and updates compatibility notes (PyTorch support now starting at \`2.8.0\` with newer versions listed). > \> > \> \Reviewed by \[Cursor Bugbot\](https://cursor.com/bugbot) for commit dae5d4f328df629c0f998555a062f834a09e8778. Bugbot is set up for automated code reviews on this repo. Configure \[here\](https://www.cursor.com/dashboard/bugbot).\ ---
**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](https://discuss.huggingface.co/u/John6666)\ **Post date:** [May 13, 2026, 11:21am UTC](https://discuss.huggingface.co/t/nvidia-rtx-pro-6000-instead-of-h200-for-zerogpu/175960/7 "2026-05-13T11:21:16Z")
Thank you! ---
**Author:** ![Downunderbunker](https://avatars.discourse-cdn.com/v4/letter/d/8baadc/32.png) [@Downunderbunker](https://discuss.huggingface.co/u/Downunderbunker)\ **Post date:** [May 14, 2026, 8:57am UTC](https://discuss.huggingface.co/t/nvidia-rtx-pro-6000-instead-of-h200-for-zerogpu/175960/8 "2026-05-14T08:57:31Z")
When’s the updates happening? I’m hanging to get generating ---
**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](https://discuss.huggingface.co/u/John6666)\ **Post date:** [May 14, 2026, 11:36pm UTC](https://discuss.huggingface.co/t/nvidia-rtx-pro-6000-instead-of-h200-for-zerogpu/175960/9 "2026-05-14T23:36:40Z")
For now, I’ve created [ZeroGPU Blackwell Runtime Drift Recovery Guide v1](https://huggingface.co/datasets/John6666/knowledge_base_md_for_rag_1/blob/main/hf_zero_gpu_blackwell_guide_20260514.md) based on the publicly available information.